Here & Now:

an IoT Design Method

Part #2

A method from Megatris Comp.

Francesco Rago

Creative Commons License Here&Now by Megatris Comp. LLC is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Summary part #1

Contextual Characteristics

• Intelligent objects introduce a new vision for strengthening communication, relationship and

business. • Each system must be able to communicate with other systems in use because integration

brings scalability, adaptability, flexibility and greater efficiency. • The key to development are the services, not the objects. • IoT on one hand and social media on the other hand are connecting the world using a new logic

of use and service. • It is clear that the world of objects connected and communicating is able to quickly transform

and greatly our needs and our habits in a positive sense. • New business must start from this new vision of technology and the world in which we live. • To be smart means to have applications that short-circuit the information by modulating key

services in digital. Programming makes use of API (Application Programming Interface) and the Agile method, in a combination that has ushered in a new era of development called DevOps.

Business Boundary Conditions

• To have a disruptive innovation in place.

• IoT is disruptive as it breaks with the past, using technological innovation to create products and / or services. (The Economist: The path to self-disruption: Nine steps of a digital transformation)

• The breaking point is the change in strategy: instead of working on established habits of the people and the progressive improvement of existing products or services, using creativity and imagination to design products and services which do not exist yet.

Why Here & Now

• Here: the app is here means just near people and satisfies their needs • Now: continuous delivery helps to generate a continuos flow of new services with users

feedback rather than a long period of waterfall production with a unique and often bad app

Testing

The Method

Here & Now

Design Your Business

Define your business

model

Gap Analysis

Architect the new solution

Design IT archtecture

Secure Architecture

ARCHITECT THE NEW SOLUTION

Design IoT architecture

• Cloud, mobile and Internet of Things need increasingly flexible, distributed and pervasive software able to adapt to the diverse needs of the interlocutors with whom they relate.

• Driver of development, according to Gartner forecasts, will be the preferences and use of services associated with the IoT from consumers.

• The increasing information flows will be managed and controlled on analysis, integration and data representation matrices or rules

• It has to strengthen the application of release rate to remain aligned with the time to market

Basics

• Agile methodologies, DevOps, Big Data Management, analytics and collaboration with internal and external stakeholders

• Each application project must become the result of a lot of work enterprise team. • Using a road map characterized by a software development that needs to include real-time

comparison moments to validate the quality of the results and, in the case, allowing to quickly change with appropriate adjustments.

• IoT generates a big growth of complexity and this entails new set of criteria and development of release. How? Developing contextual applications, liquid, intelligent and connected. This means designing software with some new cognitive artificial intelligence able to deploy applications that have some level of understanding depending on context, learn and act autonomously with respect to routine activities

4 new application strategies

LIQUID APPLICATIONS

CONNECTED APPLICATIONS

Interrelated conditions which influence an adaptive app

CONTEXTUAL APPLICATIONS

INTELLIGENT APPLICATIONS

Microservices

9/1/2016 11

The microservice architectural style is an approach to develop a single application as a suite of small services, each running in its own process and communicating with lightweight mechanisms, often an HTTP resource API.

There is a bare minimum of centralized management of these services, which may be written in different programming languages and use different data storage technologies

Here & Now

Main Steps Names

• Setting goals • Requirements analysis

• choice of technology • choice of learning approach • choice of communication approach Agent-User

• Design • Context design (Granules definition)

• Development • Test V&V • Put into baselines and transfer in production (continuous delivery) • Fine tuning

12

We will not cover all the steps but we will give a view on the main points of method.

13

DESIGN IOT ARCHTECTURE

14

15

Architecture layers

• Three layers: 1. Component behavior 2. Coordination 3. Data transfer

• Layered component model • Behavior — labeled transition systems with disjoint sets of ports • Interaction — set of interactions (interaction = set of ports) • Priorities — strict partial order on interactions

16

Agent Model

• An agent model is based on three parts: a knowledge base, a set of actions and an "engine".

• The knowledge base maintains agent's knowledge of its status, contexts status, possible transactions events-actions.

• The action is a consequence to perceive the external world and it is a to communication act with the other agents , human or non human. In particular, the actions can be divided in primitive actions (i.e., actions which cannot be decomposed in sub-actions) and complex actions (i.e., actions which are defined by the composition of other actions). The "engine" defines the behavior of the agent.

• An agent can acquire knowledge through perception and learning actions.

17

Agent PLUS AI PLUS IOT

• Internet of Things generates large volumes of data in the process of exchanging data. It would be next to impossible for humans to extract small and useful details (knowledge) from these huge sets of data since it consumes large amounts of time. That’s where AI comes in. Data generated, when combined with Artificial Intelligence allows for predictive analysis and automation thereby making life easier.

• Both the technologies, Internet of Things and Artificial Intelligence are growing rapidly. However, there are few potential threats, security is one of them. With the increase in data, the onus of protecting this valuable information from hackers is also increased. Data, in wrong hands, can prove out to be very damaging even for an individual or for an organization.

18

Design of an agent

• Describe stutus variables • Describes transactions and list actions

• Processing • Communication

• Define learning process

19

PROCESSING DESIGN

20

Behaviour, Interaction and

Priorities.

• Component-Based Design • A component is a Labelled Transition System: (describes causal rule) • Interaction model: defines atom collaboration • For a behavior B = (Q,P,→), a priority model is a strict partial order and defines conflict

resolution strategy

21

Agent example (SSB / IMCPL)

• ASSIGN EL = AGE ATT = INT VALUE = 45 ; • ASSIGN EL = HEARTRATEAVE ATT = INT VALUE 83 ; • DO

• ACQ_EVENT SHMKEY = 56743 EL = HEARTRATE ; • HEARTRATEAVE- HEARTRATE =DIFFHEARTRATE ; • GT(DIFFHEARTRATE,50)-> PUSH EVENTCODE = HRT01 & SAVE EL = ALARMHR ATT = INT VALUE = $HEARTRATE ;

• UNTIL COUNTER = <number>

22

COMMUNICATION DESIGN

23

Communication design

• User experience design (UX, UXD, UED or XD) is the process of enhancing user satisfaction by improving the usability, accessibility, and pleasure provided in the interaction between the user and the product.

• https://en.wikipedia.org/wiki/User_experience_design • https://material.google.com/

• User experience design includes elements of interaction design, information architecture, user research, and other disciplines, and is concerned with all facets of the overall experience delivered to users. Following there is a short analysis of its constituent parts.

24

Visual design

• Visual design, also commonly known as graphic design, user interface design, communication design, and visual communication, represents the aesthetics or look-and-feel of the front end of any user interface. The purpose of visual design is to use visual elements like colors, images, and symbols to convey a message to its audience.

• Information architecture • Information architecture is the art and science of structuring and organizing the

information in products and services to support usability and ”findability”. • In the context of information architecture, information is separate from both

knowledge and data, and lies nebulously between them. It is also concerned with metadata: terms used to describe and represent content objects such as documents, people, process, and organizations.

25

interaction design

• There are many key factors of interaction design: • Creating the layout of the interface • Defining interaction patterns best suited in the context • Incorporating user needs collected during user research into the

designs • Features and information that are important to the user • Interface behavior like drag-drop, selections, and mouse-over actions • Effectively communicating strengths of the system • Making the interface intuitive by building affordances • Maintaining consistency throughout the system • Usability

26

Interaction patterns as GENRES

OF COMMUNICATION

• According to Yates and Orlikowski (1992), a genre of organizational communication is a typified communicative act having a socially defined and recognized communicative purpose with regard to its audience.

• In organizations or communities, genres are enacted when members take action by drawing their knowledge tacit or explicit of genre rules that bind a particular socially recognized purpose and appropriate elements of form and substance with certain recurrent situations. A particular instance of a genre need not reflect all the rules constituting that genre, as long as it is still recognizable as partaking of that genre. For example, business letters sent via fax are still recognized as such, even though the transmission medium has changed.

• Communicative action to be identified within the relevant social community as an instance of a certain genre.

27

Interaction patterns as

GENRES OF COMMUNICATION

• Yates and Orlikowski suggest that genres are produced, reproduced, and modified by individuals through a process of structuring .

• That is, members of a community enact a genre by drawing on their knowledge of a set of genre rules, and in so doing reproduce the genre.

28

29

Genre Why What How Who When Push Generic msg An advice or

an information End User

Appears in mobile

SSB As needed

Push with answer

Need of info A request to End User

Appears in mobile

SSB As needed

Report

Report on end user request about status

Pdf with data

Appears in reports menu

EndUser At the end of periods or at the end of an activity

Email Communicate a stutus or an accomplishment

Email Sent by internet

SSB As needed

Event RD22

Data from mobile to SSB

A payload with data to process

Microservices

Mobile As a data is generated

Consequence: ZERO UI

• At Solid Conference in June 2016 ( http://conferences.oreilly.com/solid/internet-of-things-2015 ) was explained the ZERO UI.

• The possibility to interact with EndUser without screen. This means the user must afford interaction with IoT devices that can hear our words, anticipate our needs, and sense our gestures.

• What does that mean for the future of design, especially as those screens go away? This means that an AI Agent must classify context and situation and establish communications. Zero UI is the design component of all haptic, computer vision, voice control, and artificial intelligence.

• Designers need to think about what a user is trying to do right now in any possible workflow designing all possible genre with built probability depending on frequency. This will mean that a lot of our interfaces will have to become more automatic, anticipatory, and predictive.

30

Genre References

• Winograd, T. 1994. "Categories, Disciplines, and Social Coordination." Computer Supported Cooperative Work, 2, 191197.

• Yates, J., Orlikowski, W.J. and Rennecker, J. 1997. "Collaborative Genres for Collaboration: Genre Systems in Digital Media." In Proceedings of the Thirtieth Hawaii International Conference on System Sciences, (Hawaii: January).

• Norms and Forms for Work and Interaction, Wanda J. Orlikowski and JoAnne Yates,MIT Sloan School Working Paper #367194 Center for Coordination Science Techincal Report #166

31

LEARNING DESIGN

32

Learning Design

• A constructivist-based design model-offers a systematic process of learning development

• Intervention based on the needs and perspectives of Users role including the experiences and epistemology of the developer/designer

33

Learning Design

• One thing you have always to keep in mind, NNs are heuristic models. Therefore, they learn by experience, similarly to us. You cannot "insert" pure knowledge into the NN.

There are definitely a lot of decisions to be made in designing a neural net, and there is no one right answer. However, there are a few general questions that are often helpful to think about: • What are you trying to generate as an output? Draughts seems like a challenging game to play

with a neural net, because there are a lot of potential moves and the ones available change from turn to turn, but presumably you would want the output to be the next move.

• What are your inputs? This should include anything that you think would be useful in making the decision that you want the neural net to make. In the draughts example, you'd probably need to give the neural net the locations of all pieces on the board.

• Recurrent or feed-forward? Generally, unless there's a really salient reason for giving it information about what it's done in the past, it's best to use feed-forward, because the enables you to train the net with back-propagation. For draughts, for instance, you'd probably want to use a feed-forward network.

34

Learning Design

• Do you need a hidden layer? This is a harder question to know the answer to, and might require some experimentation, unless you know a lot about the high-dimensional space that your inputs occupy. Draughts is complex enough that it seems like it would require a hidden layer, but it's hard to be sure.

• Obviously, there are a lot more decisions that can/have to be made about neural net set-up, but hopefully these will get you going.

• Modeling the problem using a neural network (or any other kind of model) is a challenging problem; there is no magic bullet for this. I would recommend reading about techniques others have developed and see if you can apply those to your problem. You could start with a reference like: https://en.wikipedia.org/wiki/Types_of_artificial_neural_networks

35

A Model of an artificial neuron

• Each neuron can classify as the input unit, one summation block, one activation block and finally one result processing unit

36

37

Model of a multi layer artificial

neuron

38

Training of A Neuron

• Training of the neural networks usually entails modification of the connecting weights

• Neural networks learn from examples and exhibit some capability for generalization beyond the training data

• Testing data are used for checking the generalization • Back-propagation network is used to solve this particular problem • During training process, data are passed to the input layer and then it

passes from layer to layer maintaining the system of forward pass. Each neuron in the hidden layer receives inputs from input layer’s neurons.

39

training

• Comparing the output values with the target values errors are being calculated • Errors are minimized with the process of iteration in case of Back-propagation

40

NN example using R language

#trainingdata input data

colnames(trainingdata) <- c("Input","Output")

#Train the neural network

#Going to have 10 hidden layers

#Threshold is a numeric value specifying the threshold for the partial

net.sqrt <- neuralnet(Output~Input,trainingdata, hidden=10,

threshold=0.01)

#nn processing

testdata <- as.data.frame((5:5)^2)

net.results <- compute(net.sqrt, testdata) #Run them through the neural

network

cleanoutput <- cbind(testdata,sqrt(testdata),+

as.data.frame(net.results$net.result))

colnames(cleanoutput) <- c("Input","Expected Output","Neural Net Output")

print(cleanoutput)

Input Expected Output Neural Net Output

25 5 4.999026493

41

summary

We have seen how design Agent :

• Processing

• Communication

• Learning

42

Summary

• We have described the more innovative aspects of our method.

• Actually components are developed using our approach and our mobile app Companion makes reference to services that were developed using Here&Now

• Megatris Comp. is focused on improving productive processes using best practices and main its goal is organizational excellence in software development.

43

Slide 44

Megatris Comp. LLC We create cloud services and mobile apps to make people life easier. Our mobile apps are integrated with Megatris Cloud to sell services

and goods.

www.megatris.com 1250 Oakmead Pkwy, Sunnyvale, CA 94085, USA